Structural change of ionic association in ionic liquid/water mixtures: A high-pressure
infrared spectroscopic study
Yasuhiro Umebayashi, Jyh-Chiang Jiang, Yu-Lun Shan, Kuan-Hung Lin, Kenta Fujii, Shiro Seki, Shin-Ichi Ishiguro, Sheng Hsien Lin, and Hai-Chou Chang
Citation: The Journal of Chemical Physics 130, 124503 (2009); doi: 10.1063/1.3100099 View online: http://dx.doi.org/10.1063/1.3100099
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/130/12?ver=pdfcov Published by the AIP Publishing
Articles you may be interested in
Association structures of ionic liquid/DMSO mixtures studied by high-pressure infrared spectroscopy J. Chem. Phys. 134, 044506 (2011); 10.1063/1.3526485
Solvation and microscopic properties of ionic liquid/acetonitrile mixtures probed by high-pressure infrared spectroscopy
J. Chem. Phys. 131, 234502 (2009); 10.1063/1.3273206
Spectroscopic study of water-NaCl-benzene mixtures at high temperatures and pressures J. Chem. Phys. 123, 214504 (2005); 10.1063/1.2131061
Erratum: “Infrared study of water–benzene mixtures at high temperatures and pressures in the two- and one-phase regions” [J. Chem. Phys. 113, 1942 (2000)]
J. Chem. Phys. 113, 8390 (2000); 10.1063/1.1316005
Infrared study of water–benzene mixtures at high temperatures and pressures in the two- and one-phase regions J. Chem. Phys. 113, 1942 (2000); 10.1063/1.481998
Structural change of ionic association in ionic liquid/water mixtures:
A high-pressure infrared spectroscopic study
Yasuhiro Umebayashi,1Jyh-Chiang Jiang,2Yu-Lun Shan,3Kuan-Hung Lin,3Kenta Fujii,4 Shiro Seki,5Shin-Ichi Ishiguro,1Sheng Hsien Lin,6,7and Hai-Chou Chang3,a兲
1
Department of Chemistry, Faculty of Science, Kyushu University, Hakozaki, Higashi-ku, Fukuoko 812-8581, Japan
2Department of Chemical Engineering, National Taiwan University of Science and Technology,
Taipei 106, Taiwan
3Department of Chemistry, National Dong Hwa University, Shoufeng, Hualien 974, Taiwan
4Department of Chemistry and Applied Chemistry, Faculty of Science and Engineering, Saga University,
Honjo-machi, Saga 840-8502, Japan
5Material Science Research Laboratory, Central Research Institute of Electric Power Industry,
Komae, Tokyo 201-8511, Japan
6
Department of Applied Chemistry, National Chiao Tung University, Hsinchu 30010, Taiwan
7
Institute of Atomic and Molecular Sciences, Academia Sinica, P.O. Box 23-166, Taipei 106, Taiwan 共Received 4 December 2008; accepted 25 February 2009; published online 24 March 2009兲 High-pressure infrared measurements were carried out to observe the microscopic structures of two imidazolium-based ionic liquids, i.e., 1-ethyl-3-methylimidazolium bis共trifluoromethylsulfonyl兲amide 关EMI+共CF
3SO2兲2N−, EMI+TFSA−兴 and
1-ethyl-3-methylimidazolium bis共fluorosulfonyl兲amide 关EMI+共FSO
2兲2N−, EMI+FSA−兴. The results obtained at
ambient pressure indicate that the imidazolium C–H may exist in two different forms, i.e., isolated and network structures. As the sample of pure EMI+FSA− was compressed, the network configuration is favored with increasing pressure by debiting the isolated form. For EMI+TFSA−/H
2O mixtures, the imidazolium C–H peaks split into four bands at high pressures. The
new spectral features at approximately 3117 and 3190 cm−1, being concentration sensitive, can be
attributed to the interactions between the imidazolium C–H and water molecules. The alkyl C–H absorption exhibits a new band at approximately 3025 cm−1under high pressures. This observation
suggests the formation of a certain water structure around the alkyl C–H groups. The O–H stretching absorption reveals two types of O–H species, i.e., free O–H and bonded O–H. For EMI+TFSA−/H
2O mixtures, the compression leads to a loss of the free O–H band intensities, and
pressure somehow stabilizes the bonded O–H configurations. The results also suggest the non-negligible roles of weak hydrogen bonds in the structure of ionic liquids. © 2009 American
Institute of Physics.关DOI:10.1063/1.3100099兴
I. INTRODUCTION
In recent years, room-temperature ionic liquids have at-tracted considerable interest as green recyclable alternatives to the traditional organic solvents.1–5Interest in investigating the properties and applications of ionic liquids has been in-tensified owing to their dual nature as both salts and fluids. The most extensively studied ionic liquids are the 1-alkyl-3-methylimidazolium salts.1–3Due to the large size and confor-mational flexibility of the cations, the liquid state is thermo-dynamically favorable. This leads to small lattice enthalpies and large entropy changes that result in a melting tempera-ture around room temperatempera-ture. The unique properties of ionic liquids allow their use in several applications in chemical industry such as solvents in organic synthesis as
homoge-neous and biphasic transfer catalysts and in
electrochemistry.1,2 Ionic liquids have been employed as mixtures with other classic solvents, and this use calls for an
understanding of the ionic liquid-classic solvent interactions.6–14 Ionic liquids based on hydrophobic ions such as imidazolium derivatives and anions such as bis 共trif-luoromethylsulfonyl兲amide 共TFSA−, 关共CF
3SO2兲2N−兴兲 form
biphasic systems with water and can be used in separation processes.1,2 The bis共trifluoromethylsulfonyl兲amide anion is also called NTf2−, Tf2N−, and TFSI−in the literature. For the
extraction of organic products from aqueous media, ionic liquids with low water solubility are required. Therefore, the knowledge of the interactions of water and ionic liquids prior to their industrial applications is of primary importance. In-creasing attention has been devoted to conducting biocata-lytic transformation in ionic liquids.15The addition of water to ionic liquids is very common in biocatalytic work, and this is one of motivations for the current work.15,16
Spectroscopic methods have been widely applied to ionic liquids to probe their dynamics, structures, interactions, solvation, and transport. Measurements of the solvation re-sponses and microscopic solvent properties have resulted in the accumulation of a sizable database on solvation dynam-ics in ionic liquids.17–21 For example, several groups have
a兲Author to whom correspondence should be addressed. Electronic mail: hcchang@mail.ndhu.edu.tw. FAX: 8633570. Tel.: ⫹886-3-8633585.
0021-9606/2009/130共12兲/124503/6/$25.00 130, 124503-1 © 2009 American Institute of Physics
used the femtosecond optical heterodyne-detected Raman-induced Kerr effect spectroscopy 共OHD-RIKES兲 method to characterize the dynamics in ionic liquids.5 Since the early pioneering work based on the fluorescence behavior of C153 in ionic liquids,22several researchers have studied salvation dynamics in ionic liquids.18,20,23 Most of these studies have indicated that the time-resolvable part of the dynamics is biphasic or nonexponential in nature. Solvation dynamics in mixed solvents containing ionic liquids has also been probed.24–27Previous studies to the structure of ionic liquids have included the use of x-ray crystallography. Although the results of crystal structures are highly informative on the relative geometry changes, crystallography does not provide direct information on the local structure in the liquid state. Thus, experimental techniques, such as IR and Raman spec-troscopy, were often used to explore hydrogen-bonding structures of liquid mixtures. Over the past years the interest in pressure as an experimental variable has been growing in physicochemical studies.8,28–30The study of pressure effects reveals the intermolecular interactions of the ionic liquid at a constant kinetic energy or temperature.
The role of water in ionic liquids is complex and de-pends on the supramolecular structures of ionic liquids. By studying water-induced acceleration of ion diffusion, Schroder et al.31 proposed that imidazolium-based ionic liq-uids have polar and nonpolar regions. Alkyl imidazolium cat-ions with long enough alkyl chain length共nⱖ4兲 were char-acterized by the existence of structural organization at the nanometer scale, as reported by Triolo et al.32At high ionic liquid concentrations, ionic liquids seem to form clusters, as in the pure state, and water molecules interact with the clus-ters without interacting among themselves. Therefore, ionic liquids containing dissolved water may not be regarded as homogeneous solvents but have to be considered as nano-structured materials.33 Interactions of water dissolved in ionic liquids have been extensively studied using infrared spectroscopy. In particular, the O–H stretching modes of wa-ter are sensitive to the environment and inwa-termolecular inwa-ter- inter-actions. Based on attenuated total reflectance infrared mea-surements, researchers concluded that anions are responsible for the interactions of ionic liquids and water.34 Despite the numerous computational and experimental studies on ionic liquid/water mixtures, our knowledge of the interaction be-tween water and ionic liquids remains somewhat empirical.
Various studies have been made to elucidate the role of weak hydrogen bonds, such as C – H¯O and C–H¯X, in the structure of ionic liquids.35,36The observation of the C–H stretching vibration is one of the keys to characterize the presence of such a weak hydrogen bond and can serve as a useful probe to reflect the interactions between ionic liquids and water. One of the intriguing aspects of weak hydrogen bonds is that the C–H stretching band undergoes a blueshift when the C–H groups form weak hydrogen bonds.37–39This behavior is opposite to that of the classical hydrogen bond, and its underlying mechanism is still under debate.37–39One of the reasons for this controversy is the weakness of weak hydrogen bonds. Conclusive experimental evidence for the main origin of the blueshifts is difficult to obtain because weak hydrogen bonds usually coexist with other strong
in-teractions and are typically weak. Therefore, methods that enhance weak hydrogen bonding are crucial to provide sci-entists with a clear and unified view of this important phe-nomenon. Studies have shown the potential significance that pressure has on controlling the strength of weak hydrogen bonds.28–30 In this study, we show that high pressure is a sensitive method to probe weak hydrogen bonds in ionic liquid mixtures.
The effects of high pressure on intermolecular interac-tions have been the subject of extensive studies. Application of high pressure is the ideal tool to tune continuously the bonding properties of the materials. Pressure affects chemi-cal equilibrium, and the reaction volume is defined by the standard equation, ⌬V=−关RTln K/P兴T.40 If a system is
compressed, then the reaction will adjust to favor the com-ponents with smaller volume共⌬V⬍0兲. The pressure-induced changes in the vibrational characteristics are of particular interest. They yield important information on the bonding properties, especially with regard to the interplay of covalent and hydrogen bonding.
II. EXPERIMENTAL SECTION
Samples were prepared using
1-ethyl-3-methylimidazolium bis共trifluoromethylsulfonyl兲amide 共98.85% by HNMR, LOT1257559, Fluka兲, D2O 共99.9%,
Aldrich兲, H2O 共for chromatography, Merck兲, and
methanol-d4 共99.8% D, Cambridge Isotope兲.
1-ethyl-3-methylimidazolium bis共fluorosulfonyl兲amide of spectro-scopic grade 共Dai-ichi Kogyo Seiyaku Co. Ltd.兲 was used without further purification.41A diamond anvil cell共DAC兲 of Merril–Bassett design, having a diamond culet size of 0.6 mm, was used for generating pressures up to approximately 2 GPa. Two type-IIa diamonds were used for midinfrared measurements. The sample was contained in a 0.3-mm-diameter hole in a 0.25-mm-thick Inconel gasket mounted on the DAC. To reduce the absorbance of the samples, CaF2
crystals共prepared from a CaF2optical window兲 were placed
into the holes and compressed firmly prior to inserting the samples. A droplet of a sample filled the empty space of the entire hole of the gasket in the DAC, which was subse-quently sealed when the opposed anvils were pushed toward one another. Infrared spectra of the samples were measured on a PerkinElmer Fourier transform spectrophotometer 共model Spectrum RXI兲 equipped with a lithium tantalite midinfrared detector. The infrared beam was condensed through a 5X beam condenser onto the sample in the DAC. Typically, we chose a resolution of 4 cm−1共data point
reso-lution of 2 cm−1兲. For each spectrum, typically 1000 scans
were compiled. To remove the absorption of the diamond anvils, the absorption spectra of DAC were measured first and subtracted from those of the samples. Pressure calibra-tion follows Wong’s method.42,43The pressure dependence of screw moving distances was measured.
III. RESULTS AND DISCUSSION
Figure 1 displays infrared spectra of pure 1-ethyl-3-methylimidazolium bis共trifluoromethylsulfonyl兲a-mide共EMI+TFSA−兲 obtained under ambient pressure 共curve
124503-2 Umebayashi et al. J. Chem. Phys. 130, 124503共2009兲
a兲 and at 0.3 共curve b兲, 0.9 共curve c兲, 1.5 共curve d兲, 1.9 共curve e兲, 2.3 共curve f兲, and 2.5 GPa 共curve g兲. As indicated in Fig.
1共a兲, the aliphatic C–H modes of the methyl and ethyl groups are seen at 2951, 2970, and 2992 cm−1. The peaks at ap-proximately 3124 and 3162 cm−1 are assigned to coupled C–H stretching modes of C2– H, C4– H, and C5– H on the imidazolium cation. The imidazolium C–H stretching ab-sorption has two major peaks; close examinations reveal more structures. The nearly degenerated peaks may be attrib-uted to the perturbation of neighboring ions in the liquid state. This observation is consistent with the suggestions that the imidazolium C–H may exist in two different forms, i.e., isolated and network structures.44,45As the sample was com-pressed, that is, increasing the pressure from ambient 关Fig.
1共a兲兴 to 0.3 GPa 关Fig. 1共b兲兴, the imidazolium C–H bands
were blueshifted to 3133 and 3174 cm−1, respectively. As
the sample was further compressed关Figs.1共b兲–1共g兲兴, we also
observed a monotonic blueshift in frequency for the charac-teristic imidazolium C–H bands. Nevertheless, the pressure-induced frequency shifts of the imidazolium C–H bands are relatively small under pressures above 0.3 GPa. This may indicate a phase transition, i.e., pressure-induced solidifica-tion, above a pressure of 0.3 GPa. The blueshift may origi-nate from the overlap repulsion effect enhanced by hydro-static pressure. The pressure-enhanced C – H¯N and C – H¯F interactions may be a compensatory mechanism to provide additional stability.
Figure 2 displays infrared spectra of pure 1-ethyl-3-methylimidazolium bis共fluorosulfonyl兲amide 共EMI+FSA−兲
obtained under ambient pressure共curve a兲 and at 0.3 共curve b兲, 0.9 共curve c兲, 1.5 共curve d兲, 1.9 共curve e兲, 2.3 共curve f兲, and 2.5 GPa 共curve g兲. Figure 2共a兲 exhibits three bands at 2954, 2971, and 2990 cm−1 corresponding to alkyl C–H stretching modes, and the 3124 and 3164 cm−1bands can be
attributed to imidazolium C–H stretching vibrations. The C–H stretching modes underwent dramatic changes in their spectral profiles as the pressure was elevated to 0.3 GPa in curve b. As revealed, Fig. 2共b兲 shows the major alkyl C–H band located at 2980 cm−1, whereas the imidazolium C–H bands are shifted to 3126 and 3177 cm−1. The widths of imidazolium C–H stretching bands decrease in curve b, and this observation is likely related to local structures of the imidazolium ring. The decrease in width may be caused by the loss in intensity of those nearly degenerated bands attrib-uted to the isolated structures. In other words, the network configuration is favored with increasing pressure by debiting the isolated form in Fig. 2共b兲. Although the present experi-mental results may be explained by the network-isolated structural equilibrium, we cannot tell much about the specia-tion. Network species may be ion pairs 共or larger ion clus-ters兲, and the isolated species may mean the dissociation into free ions 共or smaller ion clusters兲. As the pressure was fur-ther elevated, the alkyl and imidazolium C–H bands were blueshifted in Figs. 2共b兲–2共g兲. The monotonic blueshift in frequency for the characteristic C–H bands 共P⬎0.3 GPa兲 suggests that the network configurations seem to be thermo-dynamically stable up to the pressure of 2.5 GPa.
In order to gain further insights on the local structures of imidazolium cations, we have studied concentration-dependent variation in the infrared spectra of EMI+TFSA−.
Figure 3 displays infrared spectra of an EMI+TFSA−/H 2O
mixture共mole fraction of EMI+TFSA−: approximately 0.84兲
obtained under ambient pressure共curve a兲 and at 0.3 共curve b兲, 0.9 共curve c兲, 1.9 共curve d兲, and 2.5 GPa 共curve e兲. As revealed in Fig. 3共a兲, the C–H bands are located at 2953, 2972, 2994, 3123, and 3163 cm−1, and the stretching modes of O–H appear at approximately 3565 and 3630 cm−1. We
FIG. 1. IR spectra of pure EMI+TFSA−under共a兲 ambient pressure and at
共b兲 0.3, 共c兲 0.9, 共d兲 1.5, 共e兲 1.9, 共f兲 2.3, and 共g兲 2.5 GPa. FIG. 2. Pressure dependence of IR spectra of pure EMIambient pressure and at共b兲 0.3, 共c兲 0.9, 共d兲 1.5, 共e兲 1.9, 共f兲 2.3, and 共g兲 2.5+FSA−under共a兲 GPa.
observe no drastic changes in the concentration dependence of the alkyl and imidazolium C–H band frequency at high concentrations of EMI+TFSA−; that is, mole fraction
共EMI+TFSA−兲ⱖ0.84 关cf. Figs.1共a兲and3共a兲兴. This behavior
may indicate a slight perturbation of local structures due to the presence of H2O at ambient pressure. According to
Cammarata et al.34 and Lopez-Pastor et al.,46 the two well-separated bands observed at 3565 and 3630 cm−1 can be
assigned to the antisymmetric 共3兲 and symmetric 共1兲
stretch vibrations of the water monomer or free O–H inter-acting with anions. As the mixture was compressed, i.e., in-creasing the pressure from ambient 关Fig. 3共a兲兴 to 0.9 GPa 关Fig.3共b兲兴, a loss of the free O–H band intensities was ob-served. The bonded O–H band appears as a broad feature at approximately 3400 cm−1in Figs.3共b兲–3共g兲. It appears that
pressure somehow stabilizes the bonded O–H conformation, as free O–H is likely switched to bonded O–H as high pres-sures are applied in Fig. 3. Evolution of the O–H spectral features in Fig.3 may arise from changes in the local struc-tures of water molecules, and the geometric properties of the hydrogen-bond network are likely perturbed as the pressure is elevated. We stress that the alkyl C–H stretching absorp-tion exhibits a new band at 3025 cm−1 associated with a weak shoulder at 3009 cm−1in Fig. 3共b兲. We notice that no
more vibration modes exist in this region, so this spectral feature located at approximately 3025 cm−1, being sensitive
to concentration and pressure dependence, may be assumed to arise from the interaction between alkyl C–H and H2O,
i.e., C – H¯O interaction. It is well known that the C–H covalent bond tends to shorten as a result of the formation of a hydrogen bond with a Lewis base.37–39 This observation
suggests the formation of a certain water structure around the alkyl C–H groups, but the details remain unclear.
The ring C–H stretching absorption shows four peaks under high pressure in Fig.3. For example, the imidazolium C–H peaks split into four bands located at 3117, 3141, 3179, and 3190 cm−1 共a weak shoulder兲 in Fig. 3共e兲. In order to learn the insight of the new spectral features revealed in the imidazolium C–H region, further pressure study on various amounts of EMI+TFSA−/H
2O provides direct evidence.
Fig-ure 3共f兲shows the IR spectra of a solution of mole fraction 共EMI+TFSA−兲=0.7 obtained under the pressure of 2.5 GPa.
See also Fig. S1 in Supplemental Material for more pressure-dependent results.47As revealed in Fig.3共f兲, the increases in intensities were observed for the bands at 3117 and 3190 cm−1. Comparing these spectra features in Fig. 3共e兲 with those in Fig.3共f兲, the 3117 and 3190 cm−1components,
being concentration sensitive, can be attributed to the inter-actions between the imidazolium C–H and water molecules. The lower 共redshift兲 frequency for the 3117 cm−1
compo-nent can be physically related to the well-known acidity of imidazolium C2– H. In the past, several models have been
proposed for the theoretical understanding of the C – H¯O interactions. When a molecule that is capable of forming blueshifting hydrogen bonding binds to a site with a suffi-ciently strong electrostatic field to dominate over the overlap effect, that molecule is predicted to display a redshifting hy-drogen bond. In this article, we present a means of looking at this issue by employing the high-pressure method. It is in-structive to point out that the shifts in C–H stretches may due to the replacement of an interaction between the C–H and an anion with an interaction with water. In other words, the redshift in C2– H appears to be the difference between two redshifts.
Figure 4 displays infrared spectra of an
EMI+TFSA−/methanol-d
4mixture having a mole fraction of
EMI+TFSA− equal to 0.2 obtained under ambient pressure
共curve a兲 and at 0.3 共curve b兲, 0.9 共curve c兲, 1.5 共curve d兲, 1.9 共curve e兲, 2.3 共curve f兲, and 2.5 GPa 共curve g兲. In order to reduce the O–H absorption intensity, we measured the infra-red spectrum in a solution of methanol-d4 共with minimal
methanol as the impurity兲, rather than methanol 共Fig.4兲. The
bonded O–H band appears as a broad feature at approxi-mately 3400 cm−1, and the shoulder at approximately
3560 cm−1 is assigned to free O–H stretching vibration in
Fig.4共a兲. This result indicates that at least two different types of O–H species were observed in Fig. 4共a兲, and the promi-nent O–H species is the bonded O–H. As the sample was compressed 关Figs.4共b兲–4共g兲兴, we observe blueshifts in fre-quency for the imidazolium C–H and redshifts in frefre-quency for the bonded O–H in Fig. 4. Nevertheless, the free O–H stretching band does not change its position with pressure. The presence of the shoulder at approximately 3560 cm−1in
Figs. 4共b兲–4共g兲 indicates that the free O–H is still stable under high pressure. This observation is remarkably different from the results of EMI+TFSA−/H
2O 共Fig. 3兲. Water
mol-ecules tend to form three-dimensional hydrogen-bonding structures, but molecules in pure methanol associate with each other to form short chains with an average chain length of five or so molecules.48 It is also known that
hydrogen-FIG. 3. IR spectra of an EMI+TFSA−/H
2O mixture 共mole fraction of EMI+TFSA−: 0.84兲 obtained under ambient pressure 共curve a兲 and at 0.3 共curve b兲, 0.9 共curve c兲, 1.9 共curve d兲, and 2.5 GPa 共curve e兲. Curve f shows the IR spectra of a solution of mole fraction共EMI+TFSA−兲=0.7 obtained at 2.5 GPa.
124503-4 Umebayashi et al. J. Chem. Phys. 130, 124503共2009兲
bond cooperativity due to concerted charge transfer can greatly enhance the strength of the individual hydrogen bonds involved in the coupling.28–30Attention has been paid to the cooperative effect with increasing cluster size.36Thus, the cluster size may be one of the reasons for the unique behavior of added methanol-d4 observed in Fig.4.
Figure 5 displays the optimized density functional theory–calculated structures of EMI+FSA− ion pairs 关Figs.
5共a兲–5共d兲兴 and ion-pair dimers 关Figs. 5共e兲 and5共f兲兴. It was known that the TFSA− anion shows similar intensities of
both the trans and cis conformers as revealed in OHD-RIKES spectra.5For the purpose of comparison with previ-ous TFSA− calculation,44
we only show the calculation re-sults of the trans conformer in Fig.5. All calculations were performed by using theGAUSSIAN 03program package.49We employed the B3LYP functional together with a standard 6-31+ Gⴱ basis set. All geometries were determined on the counterpoise 共CP兲-optimized surfaces. Energy results are shown in TableI. As illustrated in Figs.5共a兲–5共d兲, the C2– H and C4,5– H can be involved in hydrogen bonding in ion pairs. Based on energy results revealed in Table I, the ener-getically favored approach for the anion to interact with the imidazolium cation is through the formation of C2– H¯O 关Fig. 5共c兲兴 or C2– H¯N− 关Fig. 5共d兲兴. This finding is in
agreement with the observation that the C2– H group has a larger positive charge than the C4– H and C5– H groups. It is
known that cohesion in ionic liquids is strong and mostly electrostatic. Nevertheless, this study elucidates the non-negligible role of weak hydrogen bonds, such as C – H¯O and C – H¯N, in the structures of ionic liquids. As the clus-ters increase in size, the structural identification of the iso-mers is complicated by numerous different isomeric configu-rations. Only two of the optimized structures of ion-pair dimers are reported in Figs.5共e兲and5共f兲. As shown in Table
I, interaction energies of ion-pair dimers关Figs.5共e兲and5共f兲兴 are approximately 2.1 times of those of ion pairs关Figs.5共c兲 and 5共d兲兴. In other words, ion-pair dimers are only slightly favored over ion pairs for interaction-energy reasons. How-ever, ion pairs might be entropically favored over ion-pair dimers. Koddermann et al.44 calculated local minima struc-tures for ion pairs and ion-pair dimers of EMI+TFSA−. The predicted interaction energies of EMI+TFSA−共Koddermann
FIG. 4. IR spectra of an EMI+TFSA−/methanol-d
4 mixture having mole fraction of EMI+TFSA−equal to 0.2 obtained under ambient pressure共curve a兲 and at 0.3 共curve b兲, 0.9 共curve c兲, 1.5 共curve d兲, 1.9 共curve e兲, 2.3 共curve f兲, and 2.5 GPa 共curve g兲.
FIG. 5. Optimized structures of关共a兲–共d兲兴 the EMI+FSA−monomer and关共e兲 and共f兲兴 the EMI+FSA−dimer.
TABLE I. Calculated relative energies共hartree/mol兲, basis set superposition error共BSSE兲 共hartree/mol兲, and total interaction energies 共kcal/mol兲. Speciesa–c Relative energies BSSE −⌬E FSA− ⫺1351.680 449 EMIM+ ⫺344.385 267 a ⫺1696.171 386 0.002 058 65.0 b ⫺1696.171 279 0.002 512 64.7 c ⫺1696.183 195 0.002 596 72.1 d ⫺1696.182 919 0.002 801 71.8 e ⫺3392.381 702 0.006 486 153.0 f ⫺3392.382 371 0.006 773 153.2
aStructures illustrated in Fig.5. bFSA−:关共FSO
2兲2N−兴. cEMIM+:
et al., TableI兲44and EMI+FSA−complexes共this study, Table
I兲 are very similar. We also calculated the natural bond
or-bital 共NBO兲 charges of TFSA− and FSA− monomers. The
average NBO charges of F atoms in TFSA− and FSA− are
equal to ⫺0.36 and ⫺0.51, respectively. These results may indicate the non-negligible role of the C – H¯F interactions in EMI+FSA−. The stronger C – H¯F interactions in
EMI+FSA− may be one of the reasons for the remarkable
differences in the pressure-dependent results of EMI+TFSA− 共Fig.1兲 and EMI+FSA−共Fig.2兲.
IV. CONCLUSION
We have used the concentration and pressure-dependent IR techniques to monitor the structures of EMI+TFSA− and
EMI+FSA−. Our results indicate that the imidazolium C–H
may exist in two different forms, i.e., isolated and network structures at ambient pressure. Network species may be ion pairs 共or larger ion clusters兲, and the isolated species may mean the dissociation into free ions共or smaller ion clusters兲. As neat EMI+FSA−was compressed, the network
configura-tion is favored with increasing pressure. As
EMI+TFSA−/H2O mixtures were compressed, the alkyl C–H
absorption exhibits a new band at approximately 3025 cm−1 and the imidazolium C–H revealed new spectra features at approximately 3117 and 3190 cm−1. These new peaks can be attributed to the interactions between the C–H and water molecules. This observation suggests the formation of a cer-tain water structure around the alkyl and imidazolium C–H groups under high pressures. For EMI+TFSA−/H2O
mix-tures, free O–H is switched to bonded O–H as high pressures are applied. However, the free O–H is still stable under high pressures for the methanol mixtures.
ACKNOWLEDGMENTS
The authors thank the National Dong Hwa University and the National Science Council of Taiwan 共Contract No. NSC 95-2113-M-259-013-MY3兲 for financial support. This work has been financially supported by Grant-in-Aids for Scientific Research Nos. 18850017, 19003963, 19350033, and 20350037 from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
1Green Industrial Applications of Ionic Liquids, NATO Science Series, edited by R. D. Rogers, K. R. Seddon, and S. Volkov共Kluwer, Dordrecht, 2002兲.
2Ionic Liquids in Synthesis, edited by P. Wasserscheid and T. Welton 共Wiley VCH, Weinheim, 2002兲.
3H. Weingartner,Angew. Chem., Int. Ed. 47, 654共2008兲.
4C. Hardacre, J. D. Holbrey, M. Nieuwenhuyzen, and T. G. A. Youngs, Acc. Chem. Res. 40, 1146共2007兲.
5E. W. Castner, Jr., J. F. Wishart, and H. Shirota,Acc. Chem. Res. 40, 1217共2007兲.
6C. Schroder, T. Rudas, G. Neumayr, S. Benkner, and O. Steinhauser,J. Chem. Phys. 127, 234503共2007兲.
7D. Paschek, T. Koddermann, and R. Ludwig, Phys. Rev. Lett. 100, 115901共2008兲.
8H. C. Chang, J. C. Jiang, Y. C. Liou, C. H. Hung, T. Y. Lai, and S. H. Lin, J. Chem. Phys. 129, 044506共2008兲.
9M. Blesic, A. Lopes, E. Melo, Z. Petrovski, N. V. Plechkova, J. N. C. Lopes, K. R. Seddon, and L. P. N. Rebelo, J. Phys. Chem. B 112, 8645 共2008兲.
10M. G. Freire, P. J. Carvalho, R. L. Gardas, I. M. Marrucho, L. M. N. B. F. Santos, and J. A. P. Coutinho, J. Phys. Chem. B 112, 1604共2008兲. 11K. S. Mali, G. B. Dutt, and T. Mukherjee,J. Chem. Phys. 128, 054504
共2008兲.
12K. Behera, M. D. Pandey, M. Porel, and S. Pandey,J. Chem. Phys.127, 184501共2007兲.
13Y. Jeon, J. Sung, D. Kim, C. Seo, H. Cheong, Y. Ouchi, R. Ozawa, and H. Hamaguchi,J. Phys. Chem. B 112, 923共2008兲.
14T. Koddermann, C. Wertz, A. Heintz, and R. Ludwig,Angew. Chem., Int. Ed. 45, 3697共2006兲.
15R. A. Sheldon,Green Chem. 7, 267共2005兲.
16R. A. Sheldon, R. M. Lau, M. J. Sorgedrager, F. van Rantwijk, and K. R. Seddon,Green Chem. 4, 147共2002兲.
17J. H. Werner, S. N. Baker, and G. A. Baker,Analyst共Cambridge, U.K.兲
128, 786共2003兲.
18S. Arzhantsev, H. Jin, G. A. Baker, and M. Maroncelli,J. Phys. Chem. B
111, 4978共2007兲.
19S. N. Baker, G. A. Baker, and F. V. Bright,Green Chem.4, 165共2002兲. 20D. Seth, S. Sarkar, and N. Sarkar,J. Phys. Chem. B 112, 2629共2008兲. 21A. Sarkar and S. Pandey,J. Chem. Eng. Data 51, 2051共2006兲. 22R. Karmakar and A. Samanta,J. Phys. Chem. A 106, 4447共2002兲. 23P. Mukherjee, J. A. Crank, P. S. Sharma, A. B. Wijeratne, R. Adhikary, S.
Bose, D. W. Armstrong, and J. W. Petrich, J. Phys. Chem. B 112, 3390 共2008兲.
24A. Paul and A. Samanta, J. Phys. Chem. B 112, 947共2008兲.
25W. E. Gardinier, G. A. Baker, S. N. Baker, and F. V. Bright, Macromol-ecules 38, 8574共2005兲.
26D. Chakrabarty, A. Chakraborty, D. Seth, P. Hazra, and N. Sarkar,Chem. Phys. Lett. 397, 469共2004兲.
27H. V. R. Annapureddy, Z. Hu, J. Xia, and C. J. Margulis, J. Phys. Chem. B 112, 1770共2008兲.
28H. C. Chang, J. C. Jiang, C. Y. Chang, J. C. Su, C. H. Hung, Y. C. Liou, and S. H. Lin, J. Phys. Chem. B 112, 4351共2008兲.
29H. C. Chang, J. C. Jiang, J. C. Su, C. Y. Chang, and S. H. Lin,J. Phys. Chem. A 111, 9201共2007兲.
30H. C. Chang, J. C. Jiang, W. C. Tsai, G. C. Chen, and S. H. Lin,J. Phys. Chem. B 110, 3302共2006兲.
31U. Schroder, J. D. Wadhawan, R. G. Compton, F. Marken, P. A. Z. Suarez, C. S. Consorti, R. F. de Souza, and J. Dupont,New J. Chem.24,
1009共2000兲.
32A. Triolo, O. Russina, H. J. Bleif, and E. Di Cola,J. Phys. Chem. B 111, 4641共2007兲.
33W. Jiang, Y. Wang, and G. A. Voth,J. Phys. Chem. B 111, 4812共2007兲. 34L. Cammarata, S. G. Kazarian, P. A. Salter, and T. Welton,Phys. Chem.
Chem. Phys. 3, 5192共2001兲.
35A. Mele, C. D. Tran, and S. H. De Paoli Lacerda,Angew. Chem., Int. Ed.
42, 4364共2003兲.
36S. Komann, J. Thar, B. Kirchner, P. A. Hunt, and T. Welton,J. Chem. Phys. 124, 174506共2006兲.
37P. Hobza and Z. Havlas, Chem. Rev. 共Washington, D.C.兲 100, 4253 共2000兲.
38Y. L. Gu, T. Kar, and S. Scheiner,J. Am. Chem. Soc. 121, 9411共1999兲. 39A. Masunov, J. J. Dannenberg, and R. H. Contreras,J. Phys. Chem. A
105, 4737共2001兲.
40J. Jonas and A. Jonas, Annu. Rev. Biophys. Biomol. Struct. 23, 287 共1994兲.
41K. Fujii, S. Seki, S. Fukuda, R. Kanzaki, T. Takamuku, Y. Umebayashi, and S. Ishiguro, J. Phys. Chem. B 111, 12829共2007兲.
42P. T. T. Wong, D. J. Moffatt, and F. L. Baudais,Appl. Spectrosc. 39, 733 共1985兲.
43P. T. T. Wong and D. J. Moffatt,Appl. Spectrosc. 41, 1070共1987兲. 44T. Koddermann, C. Wertz, A. Heintz, and R. Ludwig,ChemPhysChem7,
1944共2006兲.
45A. Yokozeki, D. J. Kasprzak, and M. B. Shiflett, Phys. Chem. Chem. Phys. 9, 5018共2007兲.
46M. Lopez-Pastor, M. J. Ayora-Canada, M. Valcarcel, and B. Lendl, J. Phys. Chem. B 110, 10896共2006兲.
47See EPAPS Document No. E-JCPSA6-130-008913 for Fig. S1. For more information on EPAPS, see http://www.aip.org/pubservs/epaps.html. 48S. Dixit, W. C. K. Poon, and J. Crain,J. Phys.: Condens. Matter12, L323
共2000兲.
49M. J. Frish, G. W. Trucks, H. B. Schlegel et al.,
GAUSSIAN 03, Revision A.7, Gaussian, Inc., Pittsburgh, PA, 2003.
124503-6 Umebayashi et al. J. Chem. Phys. 130, 124503共2009兲